1. Conduid
  2. IoT
  3. Embedded Ui
MCP server · IoT

Embedded Ui

A lightweight, minimal-dependency embedded Web UI for any MCP Server.

37Low

Scored 6 months ago · breakdown

About Embedded Ui

Embedded Ui is an MCP server in the IoT category: a lightweight, minimal-dependency embedded Web UI for any MCP Server.

Install

Claude Code
claude mcp add embedded-ui -- npx -y mcp-embedded-ui
npx
npx -y mcp-embedded-ui

This server has no ConduID identity, so agent calls to it are not receipted. Pin the version you install and review the source before granting it credentials.

Ask AI

✦

Ask AI about Embedded Ui

Powered by Claude · Grounded in docs

I know everything about Embedded Ui. Ask me about installation, configuration, usage, or troubleshooting.

Security checks

  • ✓README presentThe repository ships documentation.

5 checks have not been run yet.

README

mcp-embedded-ui (TypeScript)

The TypeScript implementation of mcp-embedded-ui — a browser-based tool explorer for any MCP (Model Context Protocol) server.

What is this?

If you build an MCP server in TypeScript/JavaScript, your users interact with tools through raw JSON — no visual feedback, no schema browser, no quick way to test. This library adds a full browser UI to your server with zero dependencies and one function call.

┌───────────────────────────────────┐
│  Browser                          │
│  Tool list → Schema → Try it      │
└──────────────┬────────────────────┘
               │ HTTP / JSON
┌──────────────▼────────────────────┐
│  Your TypeScript MCP Server       │
│  + mcp-embedded-ui                │
│    (Node / Bun / Deno / Hono)     │
└───────────────────────────────────┘

What does the UI provide?

  • Tool list — browse all registered tools with descriptions and annotation badges
  • Schema inspector — expand any tool to view its full JSON Schema (inputSchema)
  • Try-it console — type JSON arguments, execute the tool, see results instantly
  • cURL export — copy a ready-made cURL command for any execution
  • Auth support — enter a Bearer token in the UI, sent with all requests

No build step. No CDN. No external dependencies. The entire UI is a single self-contained HTML page embedded in the package.

Install

npm install mcp-embedded-ui

Requires Node.js 18+ (or Bun/Deno with Web API support). Zero runtime dependencies.

Quick Start

Web Fetch API (Bun, Deno, Hono, Cloudflare Workers)

import { createHandler } from "mcp-embedded-ui";

const handler = createHandler(tools, handleCall, { title: "My Explorer", allowExecute: true });

// Use with any framework that supports Request/Response:
// Bun.serve({ fetch: (req) => handler(req, "/explorer") });
// Deno.serve((req) => handler(req, "/explorer"));

Node.js http

import http from "node:http";
import { createNodeHandler } from "mcp-embedded-ui";

const handle = createNodeHandler(tools, handleCall, {
  prefix: "/explorer",
  title: "My Explorer",
  allowExecute: true,
});

http.createServer(handle).listen(8000);
// Visit http://localhost:8000/explorer

Full working example

import http from "node:http";
import { createNodeHandler } from "mcp-embedded-ui";
import type { Tool, ToolCallHandler } from "mcp-embedded-ui";

// 1. Define your tools
const tools: Tool[] = [
  {
    name: "greet",
    description: "Say hello",
    inputSchema: {
      type: "object",
      properties: { name: { type: "string" } },
    },
  },
];

// 2. Define a handler: (name, args) -> [content, isError, traceId?]
const handleCall: ToolCallHandler = async (name, args) => {
  if (name === "greet") {
    return [
      [{ type: "text", text: `Hello, ${args.name ?? "world"}!` }],
      false,
      undefined,
    ];
  }
  return [[{ type: "text", text: `Unknown tool: ${name}` }], true, undefined];
};

// 3. Create and start the server
const handle = createNodeHandler(tools, handleCall, { prefix: "/explorer", allowExecute: true });
http.createServer(handle).listen(8000);

With auth hook

import type { AuthHook } from "mcp-embedded-ui";

const authHook: AuthHook = async (req, next) => {
  const token = req.headers["authorization"] ?? "";
  if (typeof token !== "string" || !token.startsWith("Bearer ")) {
    throw new Error("Unauthorized");
  }
  // Verify the token with your own logic (JWT, API key, session, etc.)
  return next();
};

// Pass authHook to enable, omit to disable
const handle = createNodeHandler(tools, handleCall, {
  prefix: "/explorer",
  allowExecute: true,
  authHook,
});

Auth only guards POST /tools/{name}/call. Discovery endpoints are always public. The UI has a built-in token input field — enter your Bearer token there and it's sent with every execution request.

The included demo (examples/node-demo.ts) uses a hardcoded Bearer demo-secret-token — the token is printed at startup so you know what to paste into the UI.

Dynamic tools

// Sync function — re-evaluated on every request
function getTools(): Tool[] {
  return registry.listTools();
}

// Async function
async function getTools(): Promise<Tool[]> {
  return await registry.asyncListTools();
}

const handler = createHandler(getTools, handleCall, { allowExecute: true });

API

Three-tier API

Function Returns Use case
createHandler(tools, handleCall, config?) (req: Request, prefix?) => Promise<Response> Bun, Deno, Hono, Cloudflare Workers
createNodeHandler(tools, handleCall, config?) (req, res) => void Node.js http.createServer
buildUIRoutes(tools, handleCall, config?) Route[] Power users — fine-grained route control

Parameters

Parameter Type Default Description
tools Tool[] | () => Tool[] | () => Promise<Tool[]> required MCP Tool objects
handleCall ToolCallHandler required async (name, args) => [content, isError, traceId?]
allowExecute boolean false Enable/disable tool execution (enforced server-side)
authHook AuthHook — Middleware: (req, next) => Promise<Response>
title string "MCP Tool Explorer" Page title (HTML-escaped automatically)
projectName string — Project name shown in footer
projectUrl string — Project URL linked in footer (requires projectName)

Auth Hook

The authHook is a middleware function that receives the request and a next function. Throw to reject with 401. The error response is always {"error": "Unauthorized"} — internal details are never leaked.

const authHook: AuthHook = async (req, next) => {
  const token = req.headers["authorization"];
  if (!token || !isValid(token)) {
    throw new Error("Bad token");
  }
  return next();
};

Auth only guards POST /tools/{name}/call. Discovery endpoints (GET /tools, GET /tools/{name}) are always public.

Endpoints

Method Path Description
GET / Self-contained HTML explorer page
GET /tools Summary list of all tools
GET /tools/{name} Full tool detail with inputSchema
POST /tools/{name}/call Execute a tool, returns MCP CallToolResult

Development

# Install dependencies
npm install

# Type check
npx tsc --noEmit

# Run tests
npx vitest run

# Run the demo (auth enabled with a demo token)
npx tsx examples/node-demo.ts
# Visit http://localhost:8000/explorer
# Paste "Bearer demo-secret-token" in the UI's token field to execute tools

Cross-Language Specification

This package implements the mcp-embedded-ui specification. The spec repo contains:

  • PROTOCOL.md — endpoint spec, data shapes, security checklist
  • explorer.html — shared HTML template (identical across all language implementations)
  • Feature specs — detailed requirements and test criteria

License

Apache-2.0

README copy captured when this server was indexed; it may lag the repository. The original is authoritative.

Questions

About Embedded Ui

How do I install Embedded Ui?

Run claude mcp add embedded-ui -- npx -y mcp-embedded-ui, then add the server to your MCP client's configuration.

Is Embedded Ui safe to use with an AI agent?

Its trust score is 37 out of 100 (low). It passes 1 of the 1 static security checks run so far. It has no ConduID identity yet, so agent calls to it are not receipted.

Is Embedded Ui still maintained?

Conduid hasn't recorded a commit date for this repository yet. Check the repository directly for recent activity.